Abstract Lung cancer is both the most common cancer worldwide and the leading cause of cancer death in the US. While radiation therapy (RT) is a highly effective treatment for many cancers, thoracic RT carries an increased risk of cardiovascular (CV) morbidity and mortality that limit critical gains in cancer control and survival. Despite the significance of this problem, we have a limited understanding of how RT results in CV toxicity, and the biologic and functional mechanisms and predictors of CV toxicity in patients. Fundamental questions include: how does RT affect mechanistic biologic and imaging markers of CV toxicity? Which cardiac radiation dose-volume parameters are associated with CV toxicity? Can baseline levels or early changes in biomarkers, imaging measures and radiation-dose volume parameters identify patients at risk of adverse CV clinical outcomes? Our preliminary data suggest thoracic RT results in inflammation, oxidative stress, microvascular dysfunction, and worse CV function in patients. We will extend these findings through detailed characterization of these pathways in a multi-center, longitudinal prospective cohort of nonsmall cell lung cancer patients from the University of Pennsylvania, Washington University, and the Brigham and Women?s Hospital treated with definitive thoracic chemoradiation for curative intent. We focus on lung cancer given the high prevalence of disease, the important role of RT in cancer control, the concomitant CV morbidity and mortality associated with RT, and the high RT doses delivered to the heart. Our overall objective is to determine if RT results in early, subclinical CV dysfunction using highly sensitive, quantitative biologic and functional measures; understand how cardiac dose-volume parameters influence these abnormalities; and develop multi-marker strategies in risk prediction. Our multi- center longitudinal cohort forms the basis of all Aims. In Aim 1, we will evaluate the changes in circulating biomarkers of CV stress, inflammation and vascular dysfunction, and to define the associations with RT dose- volume measures. In Aim 2, we will quantify RT-related changes in imaging-derived measures of CV function and perfusion, and to define the associations with RT dose-volume measures. In Aim 3, we will determine the prognostic value of biologic, imaging, and RT dose-volume measures as indicators of adverse CV outcomes. By using innovative methods in deep CV phenotyping to identify high risk individuals, we will personalize the delivery of RT and targeted cardioprotective interventions, and ultimately improve CV and overall patient outcomes. We will leverage our experiences in precision phenotyping of cancer patients undergoing cardiotoxic therapy to address a high-priority research gap in response to NIH PA 19-112.